Focus-based edge detection

• US 8,331,688 B2
• Filed: 01/07/2009
• Issued: 12/11/2012
• Est. Priority Date: 01/07/2009
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Pomerantz, Ori
• Primary Examiner(s): Akhavannik, Hadi
• Application Number:
  US12/349,848
• Publication Number:
  US 20100172572A1
• Time in Patent Office: 1,434 Days
• Field of Search: 382/199, 382/203
• US Class Current: 382/199
• CPC Class Codes: G06T 7/571 from focus

Abstract

A model generator computes a first image perimeter color difference value for each of a plurality of first pixels included in a first image that is captured using a first focal length, and selects one of the first image perimeter color difference values that exceeds a perimeter color difference threshold. Next, the model generator computes a second image perimeter color difference value for each of a plurality of second pixels included in a second image that is captured using a second focal length, and selects one of the second image perimeter color difference values that exceeds the perimeter color difference threshold. The model generator then determines that an edge is located at the first focal length by detecting that the selected first image perimeter color difference value is greater than the selected second image perimeter color difference value, and generates an image accordingly.

Claims

1. A computer-implemented method comprising:

retrieving a plurality of first pixels included in a first image that is captured using a first focal length;

computing a first image perimeter color difference value for each of the plurality of first pixels, resulting in a plurality of first image perimeter color difference values, wherein, for each of the plurality of first pixels, the computing further comprises;

computing a plurality of pixel pair color difference values for each of a plurality of perimeter pixel pairs; and

selecting a largest pixel pair color difference value from the plurality of pixel pair color difference values as the first image perimeter color difference value;

selecting one of the plurality of first image perimeter color difference values that exceeds a perimeter color difference threshold;

retrieving a plurality of second pixels included in a second image that is captured using a second focal length, the second focal length different than the first focal length;

computing a second image perimeter color difference value for each of the plurality of second pixels, resulting in a plurality of second image perimeter color difference values;

selecting one of the second image perimeter color difference values that exceeds the perimeter color difference threshold;

determining that an edge is located at the first focal length by detecting that the selected first image perimeter color difference value is greater than the selected second image perimeter color difference value; and

generating a model based upon determining that the edge is located at the first focal length.

2. The method of claim 1 further comprising:

identifying one of the plurality of first pixels that corresponds to the selected first image perimeter color difference value as a first image edge pixel; and

wherein each of the perimeter pixel pairs lie on a contributing perimeter relative to the first image edge pixel.

3. The method of claim 2 wherein the first image edge pixel'"'"'s largest pixel pair color difference value corresponds to a perpendicular pixel pair included in the plurality of pixel pairs, the method further comprising:

wherein the perpendicular pixel pair includes a first perimeter pixel and a second perimeter pixel, the first perimeter pixel receiving a first light source within a first light dependence area, and the second perimeter pixel receiving a second light source within a second light dependence area, the first light source different than the second light source.

4. The method of claim 3 further comprising:

determining that the perpendicular pixel pair is perpendicular to the edge in response to identifying the largest pixel pair color difference; and

utilizing information corresponding to determining that the perpendicular pixel pair is perpendicular to the edge during the model generation.

5. The method of claim 1 further comprising:

retrieving a plurality of third pixels included in a third image that is captured using a third focal length that is different than the first focal length and the second focal length;

computing a third image perimeter color difference value for each of the plurality of third pixels, resulting in a plurality of third image perimeter color difference values;

selecting one of the third image perimeter color difference values that exceeds the perimeter color difference threshold;

determining a number of local maximum color difference values using the selected first image color difference value, the selected second image color difference value, and the selected third image color difference value;

in response to determining only one local maximum color difference value, categorizing the edge as a simple edge; and

in response to determining two local maximum color difference values, categorizing the edge as a complex edge.

6. The method of claim 5 wherein, in response to categorizing the edge as a complex edge, the method further comprises:

determining that one of the local maximum color difference values corresponds to the first image perimeter color difference value and that another one of the local maximum color difference values corresponds to the third image perimeter color difference value; and

utilizing the first focal length and the third focal length during the model generation by including a first facet as one part of the complex edge at the first focal length and including a second facet as another part of the complex edge at the third focal length.

7. The method of claim 1 further comprising:

identifying one of the plurality of first pixels that corresponds to the selected first image perimeter color difference value as a first image edge pixel;

identifying one of the plurality of second pixels that corresponds to the selected second image perimeter color difference value as a second image edge pixel; and

wherein first image edge pixel and the second image edge pixel are at an equivalent pixel location in the first image and the second image.

8. A computer program product stored in a tangible computer readable medium, comprising functional descriptive material that, when executed by an information handling system, causes the information handling system to perform actions that include:

retrieving a plurality of first pixels included in a first image that is captured using a first focal length;

computing a first image perimeter color difference value for each of the plurality of first pixels, resulting in a plurality of first image perimeter color difference values, wherein, for each of the plurality of first pixels, the computing further comprises;

computing a plurality of pixel pair color difference values for each of a plurality of perimeter pixel pairs; and

selecting a largest pixel pair color difference value from the plurality of pixel pair color difference values as the first image perimeter color difference value;

selecting one of the plurality of first image perimeter color difference values that exceeds a perimeter color difference threshold;

retrieving a plurality of second pixels included in a second image that is captured using a second focal length, the second focal length different than the first focal length;

computing a second image perimeter color difference value for each of the plurality of second pixels, resulting in a plurality of second image perimeter color difference values;

selecting one of the second image perimeter color difference values that exceeds the perimeter color difference threshold;

determining that an edge is located at the first focal length by detecting that the selected first image perimeter color difference value is greater than the selected second image perimeter color difference value; and

generating a model based upon determining that the edge is located at the first focal length.

9. The computer program product of claim 8 wherein the information handling system further performs actions that include:

identifying one of the plurality of first pixels that corresponds to the selected first image perimeter color difference value as a first image edge pixel; and

wherein each of the perimeter pixel pairs lie on a contributing perimeter relative to the first image edge pixel.

10. The computer program product of claim 9 wherein the first image edge pixel'"'"'s largest pixel pair color difference value corresponds to a perpendicular pixel pair included in the plurality of pixel pairs, the information handling system further performing actions that include:

wherein the perpendicular pixel pair includes a first perimeter pixel and a second perimeter pixel, the first perimeter pixel receiving a first light source within a first light dependence area, and the second perimeter pixel receiving a second light source within a second light dependence area, the first light source different than the second light source.

11. The computer program product of claim 10 wherein the information handling system further performs actions that include:

determining that the perpendicular pixel pair is perpendicular to the edge in response to identifying the largest pixel pair color difference; and

utilizing information corresponding to determining that the perpendicular pixel pair is perpendicular to the edge during the model generation.

12. The computer program product of claim 8 wherein the information handling system further performs actions that include:

retrieving a plurality of third pixels included in a third image that is captured using a third focal length that is different than the first focal length and the second focal length;

computing a third image perimeter color difference value for each of the plurality of third pixels, resulting in a plurality of third image perimeter color difference values;

selecting one of the third image perimeter color difference values that exceeds the perimeter color difference threshold;

determining a number of local maximum color difference values using the selected first image color difference value, the selected second image color difference value, and the selected third image color difference value;

in response to determining only one local maximum color difference value, categorizing the edge as a simple edge; and

in response to determining two local maximum color difference values, categorizing the edge as a complex edge.

13. The computer program product of claim 12 wherein, in response to categorizing the edge as a complex edge, the information handling system further performs actions that include:

determining that one of the local maximum color difference values corresponds to the first image perimeter color difference value and that another one of the local maximum color difference values corresponds to the third image perimeter color difference value; and

utilizing the first focal length and the third focal length during the model generation by including a first facet as one part of the complex edge at the first focal length and including a second facet as another part of the complex edge at the third focal length.

14. The computer program product of claim 8 wherein the information handling system further performs actions that include:

identifying one of the plurality of first pixels that corresponds to the selected first image perimeter color difference value as a first image edge pixel;

identifying one of the plurality of second pixels that corresponds to the selected second image perimeter color difference value as a second image edge pixel; and

wherein first image edge pixel and the second image edge pixel are at an equivalent pixel location in the first image and the second image.

15. An information handling system comprising:

one or more processors;

a memory accessible by at least one of the processors;

a nonvolatile storage area accessible by at least one of the processors;

a set of instructions stored in the memory and executed by at least one of the processors in order to perform actions of;

retrieving a plurality of first pixels included in a first image that is captured using a first focal length;

computing a first image perimeter color difference value for each of the plurality of first pixels, resulting in a plurality of first image perimeter color difference values, wherein, for each of the plurality of first pixels, the computing further comprises;

computing a plurality of pixel pair color difference values for each of a plurality of perimeter pixel pairs; and

selecting a largest pixel pair color difference value from the plurality of pixel pair color difference values as the first image perimeter color difference value;

selecting one of the plurality of first image perimeter color difference values that exceeds a perimeter color difference threshold;

retrieving a plurality of second pixels included in a second image that is captured using a second focal length, the second focal length different than the first focal length;

computing a second image perimeter color difference value for each of the plurality of second pixels, resulting in a plurality of second image perimeter color difference values;

selecting one of the second image perimeter color difference values that exceeds the perimeter color difference threshold;

determining that an edge is located at the first focal length by detecting that the selected first image perimeter color difference value is greater than the selected second image perimeter color difference value; and

generating a model based upon determining that the edge is located at the first focal length.

16. The information handling system of claim 15 wherein the information handling system further performs actions that include:

identifying one of the plurality of first pixels that corresponds to the selected first image perimeter color difference value as a first image edge pixel; and

wherein each of the perimeter pixel pairs lie on a contributing perimeter relative to the first image edge pixel.

17. The information handling system of claim 16 wherein the first image edge pixel'"'"'s largest pixel pair color difference value corresponds to a perpendicular pixel pair included in the plurality of pixel pairs, the information handling system further performing actions that include:

wherein the perpendicular pixel pair includes a first perimeter pixel and a second perimeter pixel, the first perimeter pixel receiving a first light source within a first light dependence area, and the second perimeter pixel receiving a second light source within a second light dependence area, the first light source different than the second light source.

18. The information handling system of claim 17 wherein the information handling system further performs actions that include:

determining that the perpendicular pixel pair is perpendicular to the edge in response to identifying the largest pixel pair color difference; and

utilizing information corresponding to determining that the perpendicular pixel pair is perpendicular to the edge during the model generation.

19. The information handling system of claim 15 wherein the information handling system further performs actions that include:

retrieving a plurality of third pixels included in a third image that is captured using a third focal length that is different than the first focal length and the second focal length;

computing a third image perimeter color difference value for each of the plurality of third pixels, resulting in a plurality of third image perimeter color difference values;

selecting one of the third image perimeter color difference values that exceeds the perimeter color difference threshold;

determining a number of local maximum color difference values using the selected first image color difference value, the selected second image color difference value, and the selected third image color difference value;

in response to determining only one local maximum color difference value, categorizing the edge as a simple edge; and

in response to determining two local maximum color difference values, categorizing the edge as a complex edge.

20. The information handling system of claim 19 wherein, in response to categorizing the edge as a complex edge, the information handling system further performs actions that include:

determining that one of the local maximum color difference values corresponds to the first image perimeter color difference value and that another one of the local maximum color difference values corresponds to the third image perimeter color difference value; and

utilizing the first focal length and the third focal length during the model generation by including a first facet as one part of the complex edge at the first focal length and including a second facet as another part of the complex edge at the third focal length.